.upload(“product_updates_september_2023”)

Here’s a quarterly product newsletter to keep you up-to-date with all the changes in Neptune. Check what happened in the last 3 months.  

New 🎉

1. Scatter plots

If you have two metrics or parameters that you wish to compare or see how they relate to each other throughout the runs, you can now create a scatter plot. 

See an example of a scatter plot in the Neptune app.

2. Distraction-free mode 

You can now view run dashboards and compare views in distraction-free mode. 

3. Integrations & supported tools

We have 4 new integrations:

• Airflow 
• Modelbit (community-built)
• RecList (community-built)
• H2O Hydrogen Torch (community-built)

And one new guide that explains how to use Neptune with Catboost.

4. New in the documentation

Our docs team published a few tutorials that may be useful for you: 

• Reproducing a run
• Restarting a run from a checkpoint
• Overwriting logged metadata

We also gathered best practices for getting your team started with Neptune in the onboarding guide for teams. If you haven’t yet, give Neptune a try in a collaborative environment.

You have a feature request?

Let us know via Product Portal.

Improvements 🤓

Python client library

• We added support for logging of learning_rate in multi-GPU training setup. (neptune-tensorflow-keras 2.2.0)
• You can now disable Git tracking by passing git_ref=False when initializing a run. (neptune 1.5.0)
• We added a new environment variable NEPTUNE_DATA_DIRECTORY, which lets you specify where the .neptune folder should be created instead of the current working directory. (neptune 1.5.0)
• We added the list_fileset_files() function, which you can use to query metadata from FileSet fields. (neptune 1.6.0)

Web application

• You can now inspect dependencies and Git diffs in the Source code tab.
• In charts displaying multiple runs, we now identify a run more clearly when you hover over its graph (there’s a legend displaying the run ID and the color label of the line).

• You can now use the following keyboard shortcuts:
  • Toggle the side navigation – [
  • Switch between view modes – 1 / 2 / 3

Fixes 🔨

Python client library 

• We fixed an operation processor bug if the current working directory is different from the script directory. (neptune 1.4.0)
• Stringify_unsupported() now catches a broader MutableMapping class. (neptune 1.6.0)

Web application

• We fixed an issue where a field named “type” could not be displayed as a runs table column.

See other minor improvements and fixes in the changelog. 

On the roadmap 🔜

In the next months, you can expect:

• First Reports-related features,
• Ability to create rich data tables natively in Neptune.

More about .upload(“product_updates_september_2023”)

See also

Train, Track, and Deploy Your Models: Neptune + Modelbit Integration

Read more See also

.upload(“product_updates_september_2023”)

Read more See also

What is a Project in Neptune?

Read more See also

Model Training: Detectron2 + neptune.ai Integration [Example]

Read more

Explore more resources:

Content type

Case study Example dashboard Example project Product update Video Webinar

Area of interest

Computer vision NLP Reinforcement learning Tabular data Time series

Area of interest

Collaboration Comparison Debugging Evaluation Monitoring Reporting Reproducibility Tracking Versioning